Cross-reactive serum and memory B cell responses to spike ......2020/09/22 · 103 suggests that...

1

Title; 1 Cross-reactive serum and memory B cell responses to spike protein in SARS-CoV-2 2 and endemic coronavirus infection 3 4 Authors 5 Ge Song1,2,3,7, Wan-ting He1,2,3,7, Sean Callaghan1,2,3, Fabio Anzanello1,2,3, Deli Huang1, 6 James Ricketts1, Jonathan L. Torres4, Nathan Beutler1, Linghang Peng1, Sirena 7 Vargas1,2,3, Jon Cassell1,2,3, Mara Parren1, Linlin Yang1, Caroline Ignacio5, Davey M. 8 Smith5, James E. Voss1, David Nemazee1, Andrew B Ward2,3,4, Thomas Rogers1,5, 9 Dennis R. Burton1,2,3,6,8, Raiees Andrabi1,2,3,8 10 11 Affiliations 12 1Department of Immunology and Microbiology, The Scripps Research Institute, La Jolla, CA 13

92037, USA. 14 2IAVI Neutralizing Antibody Center, The Scripps Research Institute, La Jolla, CA 92037, USA 15 3Consortium for HIV/AIDS Vaccine Development (CHAVD), The Scripps Research Institute, La 16

Jolla, CA 92037, USA. 17 4Department of Integrative Structural and Computational Biology, The Scripps Research 18

Institute, La Jolla, CA 92037, USA. 19 5Division of Infectious Diseases, Department of Medicine, University of California, San Diego, La 20

Jolla, CA 92037, USA. 21 6Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology, 22

and Harvard University, Cambridge, MA 02139, USA. 23 7These authors contributed equally to this work. 24 8Corresponding author. Email: [email protected] (D.R.B.); [email protected] (R.A.). 25

26 Abstract 27 28 Pre-existing immune responses to seasonal endemic coronaviruses could have 29 profound consequences for antibody responses to SARS-CoV-2, either induced in 30 natural infection or through vaccination. Such consequences are well established 31 in the influenza and flavivirus fields. A first step to establish whether pre-existing 32 responses can impact SARS-CoV-2 infection is to understand the nature and extent 33 of cross-reactivity in humans to coronaviruses. We compared serum antibody and 34 memory B cell responses to coronavirus spike (S) proteins from pre-pandemic and 35 SARS-CoV-2 convalescent donors using a series of binding and functional assays. 36 We found weak evidence of pre-existing SARS-CoV-2 cross-reactive serum 37 antibodies in pre-pandemic donors. However, we found stronger evidence of pre-38 existing cross-reactive memory B cells that were activated on SARS-CoV-2 39 infection. Monoclonal antibodies (mAbs) isolated from the donors showed varying 40 degrees of cross-reactivity with betacoronaviruses, including SARS and endemic 41 coronaviruses. None of the cross-reactive mAbs were neutralizing except for one 42 that targeted the S2 subunit of the S protein. The results suggest that pre-existing 43 immunity to endemic coronaviruses should be considered in evaluating antibody 44 responses to SARS-CoV-2. 45

.CC-BY-NC-ND 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

The copyright holder for this preprintthis version posted September 23, 2020. ; https://doi.org/10.1101/2020.09.22.308965doi: bioRxiv preprint

https://doi.org/10.1101/2020.09.22.308965http://creativecommons.org/licenses/by-nc-nd/4.0/

2

Results and discussion 46 47 Well-known examples of pre-existing immunity to viruses influencing antibody (Ab) 48 responses to related viruses include original antigenic sin (OAS) in influenza virus 49 infections and antibody-dependent enhancement (ADE) in flavivirus infections 1-3. There 50 is considerable interest in establishing whether Ab or T cell responses to SARS-CoV-2, 51 through infection or vaccination, might be impacted by pre-existing immunity to other 52 coronaviruses, particularly the endemic coronaviruses (endemic HCoVs), namely the 53

betacoronaviruses (-HCoV), HCoV-HKU1 and HCoV-OC43, and the 54 alphacoronaviruses (α-HCoV), HCoV-NL63 and HCoV-229E, which are responsible for 55 non-severe infections such as common colds 4-8. In principle, pre-existing immune 56 perturbation effects could occur by interaction of SARS-CoV-2 with cross-reactive 57 circulating serum Abs or with B cells bearing cross-reactive B cell receptors (BCRs) or T 58 cells with cross-reactive T cell receptors (TCRs). While a number of studies have reported 59 on cross-reactive T cells and serum Abs 6,8-12, we investigate here both Ab and BCR 60 cross-reactivities. 61 62 Since individuals who have been infected with SARS-CoV-2 will generally also have been 63 infected with endemic HCoVs, we chose to compare COVID-19 and pre-pandemic donors 64 in terms of serum Abs and BCRs with specificity for the spike (S) protein. The rationale 65 was that the pre-pandemic donor cross-reactive responses could only be due to endemic 66 HCoV infection. However, the COVID-19 cohort could reveal the effects of SARS-CoV-2 67 infection on cross-reactive responses. 68 69 To assess serum Ab S-protein binding in the two cohorts, we used cell-surface and 70 recombinant soluble S proteins. First, we developed and utilized a high-throughput flow 71 cytometry-based cell surface spike binding assay (Cell-based ELISA; CELISA). COVID-72 19 convalescent sera from 36 donors showed strong reactivity to the SARS-CoV-2 spike 73 in the vast majority of infected donors (Fig. 1a, supplementary Fig. 1), somewhat lower 74 reactivity with the SARS-CoV-1 spike and much lower reactivity with the MERS-CoV spike 75 in a pattern consistent with sequence conservation between the 3 viruses. COVID sera 76 also exhibited strong cross-reactivity with endemic HCoV spikes, especially with the 77

HCoV-HKU1 and HCoV-OC43 -HCoVs (Fig. 1a). The α-HCoV- derived HCoV-NL63 78 spike was least reactive among the 4 endemic HCoVs. Next, we tested sera from a cohort 79 of 36 healthy human donors whose samples were collected pre-pandemic. The sera 80 showed minimal or no reactivity to SARS-CoV-2/CoV-1 and MERS-CoV spikes but 81 showed strong binding to the endemic HCoV spikes, especially against the HCoV-HKU1 82

and HCoV-OC43 -HCoVs (Fig. 1, supplementary Fig. 1). The results suggest that the 83 pre-pandemic sera, at least in our cohort, possess low levels of pre-existing SARS-CoV-84 2 circulating Abs. 85 86 To further investigate, we generated recombinant soluble S proteins of all 7 HCoVs using 87 a general stabilization strategy described elsewhere 13-15. ELISA showed a similar binding 88 pattern of the COVID and pre-pandemic sera as the CELISA (Fig. 1B, supplementary Fig. 89 1). The SARS-CoV-2 S specific binding of COVID sera in the two assay formats (CELISA 90 versus ELISA) correlated strongly (r = 0.92, p < 0.001) (supplementary Fig. 2), CELISA 91




3

being more sensitive overall. We also tested the neutralization of the COVID sera with 92 SARS-CoV-2 and the ID50 neutralization titers positively correlated with both binding 93 assays (CELISA (r = 0.72, p < 0.0001), ELISA (r = 0.68, p < 0.0001)) (supplementary Fig. 94 2). Overall, both CELISA and ELISA revealed binding Abs to all 7 HCoV spikes in COVID 95 sera but only to endemic HCoVs in the pre-pandemic sera. 96 97 To assess whether SARS-CoV-2 infection may impact serum Ab titers to endemic 98 HCoVs, we compared Ab titers to endemic HCoV S-protein in sera from COVID and pre-99 pandemic cohorts. Higher CELISA Ab titers to endemic HCoV-HKU1 S-protein, but not 100 for other HCoV spikes (HCoV-OC43, HCoV-NL63 and HCoV-229E) were observed in the 101 COVID cohort compared to the pre-pandemic cohort (supplementary Fig. 3). The result 102 suggests that SARS-CoV-2 infection may boost titers to the related HCoV-HKU1 spike 103 16,17. To further investigate, we divided individuals from the COVID cohort into two groups, 104 one with the higher SARS-CoV-2 spike Ab titers (AUC > 85,000) and the other with lower 105 titers (AUC < 85,000). Consistent with the above result, the COVID sera with higher 106 SARS-CoV-2 titers showed significantly higher binding to HCoV-HKU1 and HCoV-OC43 107 S-proteins compared to the low titer group (supplementary Fig. 3). The α-HCoVs HCoV-108 NL63 and HCoV-229E spike binding antibody titers were comparable between the two 109 groups and served as a control (supplementary Fig. 3). Since the two cohorts are not 110 matched in terms of a number of parameters and are of limited size, any conclusions 111 should be treated with caution. Nevertheless, it is noteworthy that SARS-CoV-2 infection 112

is apparently associated with enhanced -HCoVs S-protein Ab responses. A key question 113 is whether the enhanced responses arise from de novo B cell responses or from a recall 114 response of B cells originally activated by an endemic HCoV virus infection. 115 116 We were encouraged to look more closely at the Abs involved by Bio-Layer Interferometry 117 (BLI). Polyclonal serum antibodies were used as analytes with biotinylated S proteins 118 captured on streptavidin biosensors. Since the concentrations of the S protein specific 119 polyclonal Abs in the sera are unknown, these measurements can provide an estimate of 120 antibody dissociation off-rates (koff, which is antibody concentration independent) but not 121 binding constants 18. Slower dissociation off-rates would indicate greater affinity 122 maturation of antibodies with a given S protein 19. It is important to note that the off-rates 123 are likely associated with bivalent IgG binding (avidity) in the format used. Consistent with 124 the notion of SARS-CoV-2 infection activating a recall of cross-reactive HCoV S specific 125 Abs, the COVID sera Abs exhibited significantly slower off-rates with HCoV-HKU1 and 126 HCoV-NL63 S-proteins compared to pre-pandemic sera Abs (Fig. 2A-B, supplementary 127 Fig. 4). 128 129 Having probed serum cross-reactivity between coronaviruses, we next investigated 130 memory B cells in COVID individuals. We examined the reactivities of IgG+ memory B 131 cells in 8 select COVID donors (based on differential binding to HCoV spikes (Fig. 1) with 132

SARS-CoV-2, HCoV-HKU1 (-HCoV) and HCoV-NL63 (α-HCoV) S-proteins by flow 133 cytometry. Up to ~8% SARS-CoV-2 S-protein, ~4.3% HCoV-HKU1 S-protein and ~0.6% 134 for HCoV-NL63 S-protein-specific B cells were identified (Fig. 3B) in a frequency pattern 135 consistent with serum antibody binding titers. 136 137




4

To probe the specificities of SARS-CoV-2/endemic HCoV cross-reactive Abs, we sorted 138 single B cells for either SARS-CoV-2/HCoV-HKU-1 or SARS-CoV-2/HCoV-NL63 CoV S-139 protein double positivity. We isolated 20 S-protein-specific mAbs from 4 COVID donors, 140 CC9 (n=3), CC10 (n=3), CC36 (n=6) and CC40 (n=8) (Fig. 3C, supplementary Fig. 5) but 141 only 5 mAbs, 3 from the CC9 donor and 2 from the CC40 donor, exhibited cross-reactive 142 binding with HCoV-HKU1 spike (Fig. 3E). Two of the cross-reactive mAbs from the CC9 143 donor (CC9.1 and CC9.2) were clonally related. All 5 of the SARS-CoV-2/ HCoV-HKU-1 144

cross-reactive mAbs displayed binding to the genetically related -HCoV, HCoV-OC43, 145 spike but not to the α-HCoVs, HCoV-NL63 and HCoV-229E, spikes (Fig. 3G, 146 supplementary Fig. 6). Notably, one mAb (CC9.3) exhibited binding to 5 out of the 7 147 HCoVs, including the MERS-CoV S-protein (Fig. 3G, supplementary Fig. 6) suggesting 148

targeting of a highly conserved epitope on -HCoV spikes. One of the 4 SARS-CoV-149 2/HKU1-CoV S cross-reactive mAbs (CC40.8) showed weak cross neutralization against 150 SARS-CoV-2 and SARS-CoV-1 viruses (supplementary Fig. 6). Except for CC9.3 mAb, 151 all cross-reactive mAbs were encoded by VH3 family gene heavy chains (supplementary 152 Figs. 5 and 6) and possessed 5.6-13.2% (median = 10.4%) VH and 3.1-4.4% (median = 153 3.9%) VL nucleotide SHMs (Fig. 3D supplementary Fig. 5). 154 155 In principle, the SARS-CoV-2/HCOV-HKU1 S cross-reactive memory B cells could be 156 pre-existing in the COVID donors and show cross-reactivity with SARS-CoV-2 or originate 157 from the SARS-CoV-2 infection and show cross-reactivity with HCoV-HKU1 S protein. 158 The levels of SHM in the 5 cross-reactive mAbs listed above argue for the first 159 explanation. To gain further insight, we conducted BLI binding studies on the 3 cross-160 reactive mAbs, CC9.2, CC9.3 and CC40.8 (Fig. 4A). Both bivalent IgGs and monovalent 161 Fabs showed enhanced binding affinity to HCoV-HKU1 S-protein compared to SARS-162 CoV-2 S-protein (Fig. 4A) again consistent with the notion that the Abs (BCRs) arise from 163 a pre-existing HCoV-HKU1 S response. The serum and BCR data are then consistent. 164 The data above suggests elevated serum levels of Abs to HCoV-HKU1 S-protein in 165 COVID donors compared to pre-pandemic donors (Fig. 2A-B) is consistent with the notion 166 that SARS-CoV-2 activates B cells expressing pre-existing HCoV-HKU1 S-protein 167 specific BCRs to secrete the corresponding Abs. 168 169 One mechanism by which pre-existing cross-reactive antibodies might influence the 170 course of SARS-CoV-2 infection is ADE. Therefore, we investigated potential ADE of the 171 3 cross-reactive Abs using a SARS-CoV-2 live virus assay (Fig. 4B). Of the 3 cross-172 reactive antibodies, CC9.3 mAb showed a marginal increase (2-fold) in infection of SARS-173 CoV-2 virus in the FcγRIIa (K562) and FcγRIIb (Daudi) expressing target cells that can 174 mediate ADE. Further in vivo assessment would be needed to determine if this activity is 175 associated with any meaningful physiological effects. 176 177 To map the epitope specificities of the cross-reactive mAbs, we evaluated binding to a 178 number of fragments of the S-protein (Fig. 4C-D). Notably, all 5 of the SARS-CoV-179 2/HKU1-CoV cross-reactive mAbs failed to bind any of the S1 subunit domains or 180 subdomains, suggesting targeting to the more conserved S2 subunit. To identify the 181 cross-reactive neutralizing epitope recognized by mAb CC40.8, we conducted structural 182 studies of the antibody with the HKU1-CoV S protein. Using single particle negative stain 183




5

electron microscopy (nsEM) we observed that CC40.8 bound to the HCoV-HKU1 S trimer 184 near the bottom of the S2 domain (Fig. 4E-F). The Fab density in the 2D class averages 185 was blurry suggesting binding to a flexible surface exposed peptide. The flexibility also 186 precluded further 3D reconstruction. 187 188 Despite the requirement of double positivity in the B cell sorting, 15/20 mAbs were largely 189 specific for SARS-CoV-2. Again, like cross-reactive mAbs above, the vast majority of 190 SARS-CoV-2 specific mAbs were encoded by VH3 family gene-encoded heavy chains 191 (Fig. 3C, supplementary Fig. 5), consistent with other studies 20-26. Compared to the cross-192 reactive mAbs, the nucleotide SHM levels in SARS-CoV-2 specific mAbs were much 193 lower (VH, 0-17% (median = 0.7%) VL, 0-3.5% (median = 1.8%)) (Fig. 3D supplementary 194 Fig. 5). 3 of the 15 SARS-CoV-2 S specific mAbs showed neutralization against SARS-195 CoV-2 virus, CC40.1 being the most potent (Fig. 3F, supplementary Fig. 6). Some of the 196 SARS-CoV-2 specific mAbs exhibited cross-reactive binding with SARS-CoV-1 S protein 197 but none neutralized SARS-CoV-1. 198 199 In conclusion, using a range of immune monitoring assays, we compared the serum and 200 memory B cell responses to the S-protein from all 7 coronaviruses infecting humans in 201 SARS-CoV-2 donors and in pre-pandemic donors. In sera from our pre-pandemic cohort, 202 we found no evidence of pre-existing SARS-CoV-2 S-protein reactive antibodies that 203 resulted from endemic HCoV infections. A recent study has however reported the 204 presence of SARS-CoV-2 S-protein reactive antibodies in a small fraction of pre-205 pandemic human sera 11. An in-depth examination for the presence of SARS-CoV-2 S-206 protein reactive antibodies in large pre-pandemic human cohorts is warranted to reliably 207 determine the frequency of such antibodies. Notably, we observed serum levels of 208 endemic HCoV S-protein antibodies were higher in SARS-CoV-2-experienced donors 209 and memory B cell studies suggested these likely arose from SARS-CoV-2 infection 210 activating cross-reactive endemic HCoV S-protein-specific B cells. Cross-reactive mAbs 211 largely target the more conserved S2 subunit on S-proteins and we identified a SARS-212 CoV-2 cross-neutralizing epitope that could facilitate vaccine design and antibody-based 213 intervention strategies. Indeed, studies have shown targeting of conserved S2 subunit 214 neutralizing epitopes in SARS-CoV-2 infected donors and by SARS-CoV-1 nAbs that may 215 potentially display activities against a broader range of human coronaviruses 27-30. 216 Overall, our study highlights the need to understand fully the nature of pre-existing 217 endemic HCoV immunity in large and diverse human cohorts as vaccination of hundreds 218 of millions of people against COVID-19 is considered. 219 220 221 222




6

Acknowledgements 223 We thank all the COVID-19 cohort and healthy human cohort participants for donating 224 samples. This work was supported by the NIH CHAVD (UM1 AI44462 to A.B.W. and 225 D.R.B.), and R01 (AI132317, AI073148 to D.N.) awards, the IAVI Neutralizing Antibody 226 Center, the Bill and Melinda Gates Foundation (OPP 1170236 to A.B.W. and D.R.B.). 227 This work was also supported by the John and Mary Tu Foundation and the Pendleton 228 Trust. 229 230 Author contributions 231 R.A. and D.R.B. conceived and designed the study. T.F.R., N.B., J.R., M.P., L.Y., C.I. 232 and D.M.S. recruited donors, collected and processed plasma and PBMC samples; G.S., 233 W.H., S.C., F.A., D.H., J.R., J.L.T., N.B., L.P., S.V., and J.C. made substantial 234 contributions to the acquisition of data and data analyses; G.S., W.H., S.C., F.A., D.H., 235 J.R., J.L.T., N.B., L.P., S.V., J.C., J.E.V., D.N., A.B.W., T.F.R., D.R.B., and R.A. designed 236 experiments and analyzed the data. R.A. and D.R.B. wrote the paper and all authors 237 reviewed and edited the paper. 238 239 Competing interests 240 Competing interests: R.A., G.S., W.H., T.F.R., and D.R.B. are listed as inventors on 241 pending patent applications describing the SARS-CoV-2 and HCoV-HKU1 cross-reactive 242 antibodies. D.R.B. is a consultant for IAVI. All other authors have no competing interests 243 to declare. 244 245




7

Figure legends 246 247

248 Fig. 1. Reactivity of COVID and pre-pandemic human sera with cell surface-249 expressed human coronaviruses spikes and their soluble S-protein versions. 250 A. Heatmap showing cell-based flow cytometry binding (CELISA) of COVID and pre-251

pandemic donor sera with 293T cell surface-expressed full-length spike proteins from -252 (SARS-CoV-2, SARS-CoV-1, MERS-CoV, HCoV-HKU1, HCoV-OC43) and α-(HCoV-253 NL63 and HCoV-229E) human coronaviruses (HCoVs). Sera were titrated (6 dilutions- 254




8

starting at 1:30 dilution) and the extent of binding to cell surface-expressed HCoVs was 255 recorded by % positive cells, as detected by PE-conjugated anti-human-Fc secondary Ab 256 using flow cytometry. Area-under-the-curve (AUC) was calculated for each binding 257 titration curve and the antibody titer levels are color-coded as indicated in the key. Binding 258 of sera to vector-only plasmid (non-spike) transfected 293T cells served as a control for 259 non-specific binding. 260

B. ELISA binding of COVID and pre-pandemic donor sera to soluble S-proteins from -261

(SARS-CoV-2, SARS-CoV-1, MERS-CoV, HCoV-HKU1, HCoV-OC43) and -(HCoV-262 NL63 and HCoV-229E) HCoVs. Serum dilutions (8 dilutions- starting at 1:30 dilution) were 263 titrated against the S-proteins and the binding was detected as OD405 absorbance. AUC 264 representing the extent of binding was calculated from binding curves of COVID (left) and 265 pre-pandemic (right) sera with S-proteins and comparisons of antibody binding titers are 266 shown. Binding to BSA served as a control for non-specific binding by the sera. 267 Statistical comparisons between two groups were performed using a Mann-Whitney test, 268 (**p 0.05). 269 270




9

271

272 Fig. 2. BioLayer Interferometry binding of COVID and pre-pandemic serum 273 antibodies to SARS-CoV-2 and endemic HCoV S-proteins. 274 A. Heatmap summarizing the apparent BLI binding off-rates (koff (1/s)) of the COVID and 275

pre-pandemic human serum antibodies to SARS-CoV-2 S and endemic -HCoV, HCoV-276 HKU1 and α-HCoV, HCoV-NL63 S-proteins. Biotinylated HCoV S-proteins (100nM) were 277 captured on streptavidin biosensors to achieve binding of at least 1 response unit. The S-278 protein-immobilized biosensors were immersed in 1:40 serum dilution solution with serum 279 antibodies as the analyte and the association (120 s; 180-300) and dissociation (240 s; 280 300-540) steps were conducted to detect the kinetics of antibody-protein interaction. koff 281 (1/s) dissociation rates for each antibody-antigen interaction are shown. 282 B. Off-rates for binding of serum antibodies from COVID donors and from pre-pandemic 283 donors to SARS-CoV-2 S and endemic HCoV, HCoV-HKU1 and HCoV-NL63, S proteins. 284 Significantly lower dissociation off-rates are observed for COVID compared to pre-285 pandemic sera. Statistical comparisons between the two groups were performed using a 286 Mann-Whitney test. 287 288




10

289 Fig. 3. SARS-CoV-2 S and endemic HCoV S-protein specific cross-reactive IgG+ 290 memory B cells from COVID donors and isolation of and characterization of mAbs. 291 A-B. Flow cytometry analysis showing the single B cell sorting strategy for COVID 292

representative donor CC9 and frequencies of SARS-CoV-2 S and endemic -HCoV, 293 HCoV-HKU1 and α-HCoV, HCoV-NL63 S-protein specific memory B cells in 8 select 294 COVID donors. The B cells were gated as SSL, CD4-, CD8-, CD11C-, IgD-, IgM-, CD19+, 295 IgG+. The frequencies of HCoV S-protein-specific IgG memory B cells were as follows; 296 SARS-CoV-2 S (up to ~8% - range = ~1.6-8%), HCoV-HKU1 S (up to ~4.3% - range = 297 ~0.2-4.3%), HCoV-NL63 S (up to ~0.6% - range = ~0.04-0.6%) protein single positive 298 and SARS-CoV-2/HCoV-HKU1 S (up to ~2.4% - range = ~0.02-2.4%) and SARS-CoV-299 2/HCoV-NL63 S-protein (up to ~0.09% - range = ~0-0.09%) double positives. SARS-CoV-300 2 infected donors showed the presence of SARS-CoV-2/HCoV-HKU1 S-protein cross-301 reactive IgG memory B cells. A Mann-Whitney test was used to compare the levels of 302




11

HCoV S-protein specific IgG memory B cells and the p-values for each comparison are 303 indicated. **p

12

322 Fig. 4. Binding, ADE and epitope specificities of SARS-CoV-2/HCoV-HKU1 S-323 protein specific cross-reactive mAbs. 324 A. BLI of SARS-CoV-2 and HCoV-HKU1 S-protein-specific cross-reactive mAbs. BLI 325 binding of both IgG and Fab versions of 3 cross-reactive mAbs (CC9.2, CC9.3 and 326 CC40.8) to SARS-CoV-2 and HCoV-HKU1 S-proteins was tested and the binding curves 327 show association (120 s; 180-300) and dissociation rates (240 s; 300-540). BLI binding 328 of antibody-S-protein combinations shows more stable binding (higher binding constants 329 (KDs)) of cross-reactive mAbs HCoV-HKU1 compared to the SARS-CoV-2 S protein. 330 B. Antibody Dependent Enhancement (ADE) activities of cross-reactive mAbs, CC9.2, 331 CC9.3 and CC40.8 bonding to SARS-CoV-2 live virus using FcγRIIa (K562) and FcγRIIb 332 (Daudi)-expressing target cells. A dengue antibody, DEN3, was used as a control. 333 C-D. Epitope mapping of the mAbs binding to domains and subdomains of SARS-CoV-2 334 S-protein, NTD, RBD, RBD-SD1 and RBD-SD1-2 and heatmap showing BLI responses 335




13

for each protein. The extent of binding responses is color coded. 5 mAbs were specific 336 for RBD, 2 for NTD and the remaining mAbs displayed binding only to the whole S protein. 337 E-F. Negative stain electron microscopy of HCoV-HKU1 S-protein + Fab CC40.8 338

complex and comparison to MERS-CoV S + Fab G4 complex. (E) Raw micrograph of 339

HCoV-HKU1 S in complex with Fab CC40.8. (F) Select reference-free 2D class 340

averages with Fabs colored in orange for Fab CC40.8 and blue for Fab G4, which in 2D 341

appear to bind a proximal epitope at the base of the trimer. 2D projections for MERS-342

CoV S-protein in complex with Fab G4 were generated in EMAN2 from PDB 5W9J. 343




14

Methods 344 345 Plasmid construction for full-length and recombinant soluble proteins 346 To generate full-length human coronavirus plasmids, the spike genes were synthesized 347 by GeneArt (Life Technologies). The SARS-CoV-1 (1255 amino acids; GenBank: 348 AAP13567), SARS-CoV-2 (1273 amino acids; GenBank: MN908947), MERS-CoV (1353 349 amino acids; GenBank: APB87319.1), HCoV-HKU1 (1356 amino acids; GenBank: 350 YP_173238.1), HCoV-OC43 (1361 amino acids; GenBank: AAX84792.1), HCoV-NL63 351 (1356 amino acids; GenBank: YP_003767.1) and HCoV-229E (1173 amino acids; 352 GenBank: NP_073551.1) were cloned into the mammalian expression vector phCMV3 353 (Genlantis, USA) using PstI and BamH restriction sites. To express the soluble S 354 ectodomain protein SARS-CoV-1 (residue 1-1190), SARS-CoV-2 (residue 1-1208), 355 MERS-CoV (residue 1-1291), HCoV-HKU1 (residue 1-1295), HCoV-OC43 (residue 1-356 1300) and HCoV-NL63 (residue 1-1291), HCoV-229E (residue 1-1110), the 357 corresponding DNA fragments were PCR amplified and constructed into vector phCMV3 358 using a Gibson assembly kit. To trimerize the soluble S proteins and stabilize them in the 359 prefusion state, we incorporated a C-terminal T4 fibritin trimerization motif in the C-360 terminal of each constructs and two consecutive proline substitutions in the S2 subunit 13-361 15. To be specific, the K968/V969 in SARS-CoV-1, the K986/V987 in SARS-CoV-2, the 362 V1060/L1061 in MERS-CoV, the A1071/L1072 in HCoV-HKU1, the A1078/L1079 in 363 HCoV-OC43, the S1052/I1053 in HCoV-NL63 and the T871/I872 in HCoV-229E were 364 replaced by proline residues. Additionally, the S2 cleavage sites in each protein were 365 replaced with a “GSAS” linker peptide. To facilitate the purification and biotin labeling of 366 the soluble protein, the HRV-3C protease cleavage site, 6X HisTag, and AviTag spaced 367 by GS-linkers were added to the C-terminus of the constructs, as needed. To express the 368 SARS-CoV-2 N-terminal domain-NTD (residue 1-290), receptor-binding domain-RBD 369 (residue 320-527), RBD-SD1 (residue 320-591), and RBD-SD1-2 (residue 320-681) 370 subdomains, we amplified the DNA fragments by PCR reaction using the SARS-CoV-2 371 plasmid as template. All the DNA fragments were cloned into the vector phCMV3 372 (Genlantis, USA) in frame with the original secretion signal or the Tissue Plasminogen 373 Activator (TPA) leader sequence. All the truncation proteins were fused to the C-terminal 374 6X HisTag, and AviTag spaced by GS-linkers to aid protein purification and biotinylation. 375 376 Expression and purification of the proteins 377 To express the soluble S ectodomain proteins of each human coronavirus and the 378 truncated versions, the plasmids were transfected into FreeStyle293F cells (Thermo 379 Fisher). For general production, 350 ug plasmids were transfected into 1L FreeStyle293F 380 cells at the density of 1 million cells/mL. We mixed 350 ug plasmids with 16mL 381 transfectagro™ (Corning) and 1.8 mL 40K PEI (1mg/mL) with 16mL transfectagro™ in 382 separate 50 mL conical tubes. We filtered the plasmid mixture with 0.22 μm Steriflip™ 383 Sterile Disposable Vacuum Filter Units (MilliporeSigma™) before combining it with the 384 PEI mixture. After gently mixing the two components, the combined solution rested at 385 room temperature for 30 min and was poured into 1 L FreeStyle293F cell culture. To 386 harvest the soluble proteins, the cell cultures were centrifuged at 3500 rpm for 15 min on 387 day 4 after transfection. The supernatants were filtered through the 0.22 μm membrane 388 and stored in a glass bottle at 4 ℃ before purification. The His-tagged proteins were 389




15

purified with the HisPur Ni-NTA Resin (Thermo Fisher). To eliminate nonspecific binding 390 proteins, each column was washed with at least 3 bed volumes of wash buffer (25 mM 391 Imidazole, pH 7.4). To elute the purified proteins from the column, we loaded 25 mL of 392 the elution buffer (250 mM Imidazole, pH 7.4) at slow gravity speed (~4 sec/drop). 393 Proteins without His tags were purified with GNL columns (Vector Labs). The bound 394 proteins were washed with PBS and then eluted with 50 mL of 1M Methyl α-D-395 mannopyranoside (Sigma M6882-500G) in PBS. By using Amicon tubes, we buffer 396 exchanged the solution with PBS and concentrated the proteins. The proteins were 397 further purified by size-exclusion chromatography using a Superdex 200 Increase 10/300 398 GL column (GE Healthcare). The selected fractions were pooled and concentrated again 399 for further use. 400 401 Biotinylation of proteins 402 Random biotinylation of S proteins was conducted using EZ-Link NHS-PEG Solid-Phase 403 Biotinylation Kit (Thermo Scientific #21440). 10ul DMSO were added per tube for making 404 concentrated biotin stock, 1ul of which were diluted into 170ul water before use. 405 Coronavirus spike proteins were concentrated to 7-9 mg/ml using 100K Amicon tubes in 406 PBS, then aliquoted into 30ul in PCR tubes. 3ul of the diluted biotin were added into each 407 aliquot of concentrated protein and incubated on ice for 3h. After reaction, buffer 408 exchange for the protein was performed using PBS to remove excess biotin. BirA 409 biotinylation of S proteins was conducted using BirA biotin-protein ligase bulk reaction kit 410 (Avidity). Coronavirus S proteins with Avi-tags were concentrated to 7-9 mg/ml using 411 100K Amicon tubes in TBS, then aliquoted into 50ul in PCR tubes. 7.5ul of BioB Mix, 7.5ul 412 of Biotin200, and 5ul of BirA ligase (3mg/ml) were added per tube. The mixture was 413 incubated on ice for 3h, followed by size-exclusion chromatography to segregate the 414 biotinylated protein and the excess biotin. The extend of biotinylation was evaluated by 415 BioLayer Interferometry binding value using streptavidin biosensors. 416 417 CELISA binding 418 Binding of serum antibodies or mAbs to human coronavirus spike proteins expressed on 419 HEK293T cell surface was determined by flow cytometry, as described previously 31. 420 HEK293T cells were transfected with plasmids encoding full-length coronavirus spikes 421 including SARS-CoV-1, SARS-CoV-2, MERS-CoV, HCoV-HKU1, HCoV-OC43, HCoV-422 NL63 and HCoV-229E. Transfected cells were incubated for 36-48 h at 37°C. Post 423 incubation cells were trypsinized to prepare a single cell suspension and were distributed 424 into 96-well plates. Serum samples were prepared as 3-fold serial titrations in FACS 425 buffer (1x PBS, 2% FBS, 1 mM EDTA), starting at 1:30 dilution, 6 dilutions. 50 μl/well of 426 the diluted samples were added into the cells and incubated on ice for 1h. The plates 427 were washed twice in FACS buffer and stained with 50 μl/well of 1:200 dilution of R-428 phycoerythrin (PE)-conjugated mouse anti-human IgG Fc antibody (SouthernBiotech 429 #9040-09) and 1:1000 dilution of Zombie-NIR viability dye (BioLegend) on ice in dark for 430 45min. After another two washes, stained cells were analyzed using flow cytometry (BD 431 Lyrics cytometers), and the binding data were generated by calculating the percent (%) 432 PE-positive cells for antigen binding using FlowJo 10 software. CR3022, a SARS-CoV-1 433 and SARS-CoV-2 spike binding antibody, and dengue antibody, DEN3, were used as 434 positive and negative controls for the assay, respectively. 435




16

436 ELISA binding 437 96-well half-area plates (Corning cat. #3690, Thermo Fisher Scientific) were coated 438 overnight at 4°C with 2 ug/ml of mouse anti-His-tag antibody (Invitrogen cat. #MA1-439 21315-1MG, Thermo Fisher Scientific) in PBS. Plates were washed 3 times with PBS plus 440 0.05% Tween20 (PBST) and blocked with 3% (wt/vol) bovine serum albumin (BSA) in 441 PBS for 1 h. After removal of the blocking buffer, the plates were incubated with His-442 tagged spike proteins at a concentration of 5 ug/ml in 1% BSA plus PBS-T for 1.5 hr at 443 room temperature. After a washing step, perturbed and lotus serum samples were added 444 in 3-fold serial dilutions in 1% BSA/PBS-T starting from 1:30 and 1:40 dilution, 445 respectively, and incubated for 1.5 hr. CR3022 and DEN3 human antibodies were used 446 as a positive and negative control, respectively, and added in 3-fold serial dilutions in 1% 447 BSA/PBS-T starting at 10 ug/ml. After the washes, a secondary antibody conjugated with 448 alkaline phosphatase (AffiniPure goat anti-human IgG Fc fragment specific, Jackson 449 ImmunoResearch Laboratories cat. #109-055-008) diluted 1:1000 in 1% BSA/PBS-T, 450 was added to each well. After 1 h of incubation, the plates were washed and developed 451 using alkaline phosphatase substrate pNPP tablets (Sigma cat. #S0942-200TAB) 452 dissolved in a stain buffer. The absorbance was measured after 8, 20, and 30 minutes, 453 and was recorded at an optical density of 405 nm (OD405) using a VersaMax microplate 454 reader (Molecular Devices), where data were collected using SoftMax software version 455 5.4. The wells without the addition of serum served as a background control. 456 457 BioLayer Interferometry binding 458 An Octet K2 system (ForteBio) was used for performing the binding experiments of the 459 coronavirus spike proteins with serum samples. All serum samples were prepared in 460 Octet buffer (PBS plus 0.1% Tween20) as 1:40 dilution, random-biotinylated S proteins 461 were prepared at a concentration of 100nM. The hydrated streptavidin biosensors 462 (ForteBio) first captured the biotinylated spike proteins for 60s, then transferred into Octet 463 buffer for 60s to remove unbound protein and provide the baseline. Then, they were 464 immersed in diluted serum samples for 120s to provide association signal, followed by 465 transferring into Octet buffer to test for disassociation signal for 240s. The data generated 466 was analyzed using the ForteBio Data Analysis software for correction and curve fitting, 467 and for calculating the antibody dissociation rates (koff values) or KD values for 468 monoclonal antibodies. 469 470 Flow cytometry B cell profiling and mAb isolation with HCoV S proteins 471 Flow cytometry of PBMC samples from convalescent human donors were conducted 472 following methods described previously 22,32,33. Frozen human PBMCs were re-473 suspended in 10 ml RPMI 1640 medium (Thermo Fisher Scientific, #11875085) pre-474 warmed to 37°C containing 50% fetal bovine serum (FBS). After centrifugation at 400 x g 475 for 5 minutes, the cells were resuspended in a 5 ml FACS buffer (PBS, 2% FBS, 2mM 476 EDTA) and counted. A mixture of fluorescently labeled antibodies to cell surface markers 477 was prepared, including antibodies specific for the T cell markers CD3(APC Cy7, BD 478 Pharmingen #557757), CD4(APC-Cy7, Biolegend #317418) and CD8(APC-Cy7, BD 479 Pharmingen #557760); B cell markers CD19 (PerCP-Cy5.5, Fisher Scientific 480 #NC9963455), IgG(BV605, BD Pharmingen #563246) and IgM(PE); CD14(APC-Cy7, BD 481




17

Pharmingen #561384, clone M5E2). The cells were incubated with the antibody mixture 482 for 15 minutes on ice in the dark. The SARS-CoV-2 S protein was conjugated to 483 streptavidin-AF488 (Life Technologies #S11223), the HCoV-HKU1 S protein to 484 streptavidin-BV421 (BD Pharmingen #563259) and the HCoV-NL63 S protein to 485 streptavidin-AF647 (Life Technologies #S21374). Following conjugation, each S protein-486 probe was added to the Ab-cell mixture and incubated for 30 minutes on ice in the dark. 487 FVS510 Live/Dead stain (Thermo Fisher Scientific, #L34966) in the FACS buffer (1:300) 488 was added to the cells and incubated on ice in the dark for 15 minutes. The stained cells 489 were washed with FACS buffer and re-suspended in 500 μl of FACS buffer/10-20 million 490 cells, passed through a 70 μm mesh cap FACS tube (Fisher Scientific, #08-771-23) and 491 sorted using a Beckman Coulter Astrios sorter, where memory B cells specific to S protein 492 proteins were isolated. In brief, after the gating of lymphocytes (SSC-A vs. FSC-A) and 493 singlets (FSC-H vs. FSC-A), live cells were identified by the negative FVS510 Live/Dead 494 staining phenotype, then antigen-specific memory B cells were distinguished with 495 sequential gating and defined as CD3-, CD4-, CD8-, CD14-, CD19+, IgM-and IgG+. 496 Subsequently, the S protein specific B cells were identified with the phenotype of 497 AF488+BV421+ (SARS-CoV-2/HCoV-HKU1 S protein double positive) or 498 AF488+AF647+ (SARS-CoV-2/HCoV-NL63 S protein double positive). Positive memory 499 B cells were then sorted and collected at single cell density in 96-well plates. Downstream 500 single cell IgG RT-PCR reactions were conducted using Superscript IV Reverse 501 Transcriptase (Thermo Fisher, # 18090050), random hexamers (Gene Link # 26400003), 502 Ig gene-specific primers, dNTP, Igepal, DTT and RNAseOUT (Thermo Fisher # 503 10777019). cDNA products were then used in nested PCR for heavy/light chain variable 504 region amplification with HotStarTaq Plus DNA Polymerase (QIAGEN # 203643) and 505 specific primer sets described previously 34,35. The second round PCR exploited primer 506 sets for adding on the overlapping region with the expression vector, followed by cloning 507 of the amplified variable regions into vectors containing constant regions of IgG1, Ig 508 Kappa, or Ig Lambda using Gibson assembly enzyme mix (New England Biolabs 509 #E2621L) after confirming paired amplified product on 96-well E gel (ThermoFisher 510 #G720801). Gibson assembly products were finally transformed into competent E.coli 511 cells and single colonies were picked for sequencing and analysis on IMGT V-Quest 512 online tool (http://www.imgt.org) as well as downstream plasmid production for antibody 513 expression. 514 515 Neutralization assay 516 Under BSL2/3 conditions, MLV-gag/pol and MLV-CMV plasmids were co-transfected into 517 HEK293T cells along with full-length or variously truncated SARS-CoV1 and SARS-COV2 518 spike plasmids using Lipofectamine 2000 to produce single-round of infection competent 519 pseudo-viruses. The medium was changed 16 hours post transfection. The supernatant 520 containing MLV-pseudotyped viral particles was collected 48h post transfection, aliquoted 521 and frozen at -80 °C for neutralization assay. Pseudotyped viral neutralization assay was 522 performed as previously described with minor modification (Modified from TZM-bl assay 523 protocol 36). 293T cells were plated in advance overnight with DMEM medium +10% FBS 524 + 1% Pen/Strep + 1% L-glutamine. Transfection was done with Opti-MEM transfection 525 medium (Gibco, 31985) using Lipofectamine 2000. The medium was changed 12 hours 526 after transfection. Supernatants containing the viruses were harvested 48h after 527




18

transfection. 1) Neutralization assay for plasma. plasma from COVID donors were heat-528 inactivated at 56°C for 30 minutes. In sterile 96-well half-area plates, 25μl of virus was 529 immediately mixed with 25 μl of serially diluted (3x) plasma starting at 1:10 dilution and 530 incubated for one hour at 37°C to allow for antibody neutralization of the pseudotyped 531 virus. 10,000 HeLa-hACE2 cells/ well (in 50ul of media containing 20μg/ml Dextran) were 532 directly added to the antibody virus mixture. Plates were incubated at 37°C for 42 to 48 533 h. Following the infection, HeLa-hACE2 cells were lysed using 1x luciferase lysis buffer 534 (25mM Gly-Gly pH 7.8, 15mM MgSO4, 4mM EGTA, 1% Triton X-100). Luciferase 535 intensity was then read on a Luminometer with luciferase substrate according to the 536 manufacturer’s instructions (Promega, PR-E2620). 2) Neutralization assay for 537 monoclonal antibodies. In 96-well half-area plates, 25ul of virus was added to 25ul of five-538 fold serially diluted mAb (starting concentration of 50ug/ml) and incubated for one hour 539 before adding HeLa-ACE2 cell as mentioned above. Percentage of neutralization was 540 calculated using the following equation: 100 X (1 – (MFI of sample – average MFI of 541 background) / average of MFI of probe alone – average MFI of background)). 542 543 Antibody dependent enhancement assay 544 Ex vivo antibody dependent enhancement (ADE) quantification was measured using a 545 focus reduction neutralization assay. Monoclonal antibodies were serially diluted in 546 complete RPMI and incubated for 1 hour at 37°C with SARS-CoV-2 strain USA-547 WA1/2020 (BEI Resources NR- 52281) [MOI=.01], in a BSL3 facility. Following the initial 548 incubation, the mAb-virus complex was added in triplicate to 384-well plates seeded with 549 1E4 of K562 or Daudi cells and were incubated at 34°C for 24 hours. 20µL of the 550 supernatant was transferred to a 384-well plate seeded with 2E3 HeLa-ACE2 cells and 551 incubated for an additional 24 hours at 34°C. Plates were fixed with 25 ul of 8% 552 formaldehyde for 1 hour at 34°C. Plates were washed 3 times with 1xPBS 0.05% Tween-553 20 following fixation. 10µL of human polyclonal sera diluted 1:500 in Perm/Wash Buffer 554 (BD Biosciences)was added to the plate and incubated at RT for 2 hours. The plates were 555 then washed 3 times with 1xPBS 0.05% Tween-20 and stained with peroxidase goat anti-556 human Fab (Jackson Scientific, 109-035-006) diluted 1:2000 in Perm/wash buffer then 557 incubated at RT for 2 hours. The plates were then washed 3 times with 1xPBS 0.05% 558 Tween-20. 10µL of Perm/Wash buffer was added to the plate then incubated for 15 559 minutes at RT. The Perm/Wash buffer was removed and 10µL of TrueBlue peroxidase 560 substrate (KPL) was added. The plates were incubated for 30 minutes at RT then washed 561 once with milli-Q water. The FFU per well was then quantified using a compound 562 microscope. The PFU/mL of the monocyte plate supernatant was calculated and graphed 563 using Prism 8 software. 564 565 Negative Stain Electron Microscopy 566 The HCoV-HKU1 S protein was incubated with a 3-fold molar excess of Fab CC40.8 for 567 30 mins at room temperature and diluted to 0.03 mg/ml in 1X TBS pH 7.4. 3 μL of the 568 diluted sample was deposited on a glow discharged copper mesh grid, blotted off, and 569 stained for 55 seconds with 2% uranyl formate. Proper stain thickness and particle density 570 was assessed on a FEI Morgagni (80keV). The Leginon software 37 was used to automate 571 data collection on a FEI Tecnai Spirit (120keV), paired a FEI Eagle 4k x 4k camera. The 572 following parameters were used: 52,000x magnification, -1.5 μm defocus, a pixel size of 573




19

2.06 Å, and a dose of 25 e−/Å2. Micrographs were stored in the Appion database 38, 574 particles were picked using DogPicker 39, and a particle stack of 256 pixels was made. 575 RELION 3.0 40 was used to generate the 2D class averages. The flexibility of the fab 576 relative to the spike precluded 3D reconstruction. 577 578 Statistical Analysis 579 Statistical analysis was performed using Graph Pad Prism 8 for Mac, Graph Pad 580 Software, San Diego, California, USA. Median area-under-the-curve (AUC) or reciprocal 581 50% binding (ID50) or neutralization (IC50) titers were compared using the non-582 parametric unpaired Mann-Whitney-U test. The correlation between two groups was 583 determined by Spearman rank test. Data were considered statistically significant at * p < 584 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001. 585 586




20

Data availability 587 The authors declare that the data supporting the findings of this study are available within 588 the paper and its supplementary information files or from the corresponding author upon 589 reasonable request. Antibody sequences have been deposited in GenBank under 590 accession numbers XXX-XXX. Antibody plasmids are available from Dennis Burton under 591 an MTA from The Scripps Research Institute. 592




21

Reference: 593 594

1. Angeletti, D., et al. Defining B cell immunodominance to viruses. Nature immunology 18, 595 456-463 (2017). 596

2. Arvin, A.M., et al. A perspective on potential antibody-dependent enhancement of 597 SARS-CoV-2. Nature 584, 353-363 (2020). 598

3. Halstead, S.B. Neutralization and antibody-dependent enhancement of dengue viruses. 599 Adv Virus Res 60, 421-467 (2003). 600

4. Che, X.Y., et al. Antigenic cross-reactivity between severe acute respiratory syndrome-601 associated coronavirus and human coronaviruses 229E and OC43. J Infect Dis 191, 602 2033-2037 (2005). 603

5. Huang, A.T., et al. A systematic review of antibody mediated immunity to coronaviruses: 604 antibody kinetics, correlates of protection, and association of antibody responses with 605 severity of disease. medRxiv (2020). 606

6. Mateus, J., et al. Selective and cross-reactive SARS-CoV-2 T cell epitopes in unexposed 607 humans. Science (2020). 608

7. Paules, C.I., Marston, H.D. & Fauci, A.S. Coronavirus Infections-More Than Just the 609 Common Cold. JAMA 323, 707-708 (2020). 610

8. Sekine, T., et al. Robust T cell immunity in convalescent individuals with asymptomatic 611 or mild COVID-19. Cell (2020). 612

9. de Assis, R.R., et al. Analysis of SARS-CoV-2 Antibodies in COVID-19 Convalescent 613 Plasma using a Coronavirus Antigen Microarray. bioRxiv (2020). 614

10. Grifoni, A., et al. Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans 615 with COVID-19 Disease and Unexposed Individuals. Cell 181, 1489-1501 e1415 (2020). 616

11. Ng, K.W., et al. Pre-existing and de novo humoral immunity to SARS-CoV-2 in humans. 617 bioRxiv, 2020.2005.2014.095414 (2020). 618

12. Prévost, J., et al. Cross-sectional evaluation of humoral responses against SARS-CoV-2 619 Spike. bioRxiv, 2020.2006.2008.140244 (2020). 620

13. Kirchdoerfer, R.N., et al. Pre-fusion structure of a human coronavirus spike protein. 621 Nature 531, 118-121 (2016). 622

14. Pallesen, J., et al. Immunogenicity and structures of a rationally designed prefusion 623 MERS-CoV spike antigen. Proceedings of the National Academy of Sciences of the 624 United States of America 114, E7348-E7357 (2017). 625

15. Wrapp, D., et al. Cryo-EM structure of the 2019-nCoV spike in the prefusion 626 conformation. Science 367, 1260-1263 (2020). 627

16. Chen, Y., Liu, Q. & Guo, D. Emerging coronaviruses: Genome structure, replication, and 628 pathogenesis. J Med Virol 92, 418-423 (2020). 629

17. Lu, R., et al. Genomic characterisation and epidemiology of 2019 novel coronavirus: 630 implications for virus origins and receptor binding. Lancet 395, 565-574 (2020). 631

18. Dennison, S.M., et al. Qualified Biolayer Interferometry Avidity Measurements 632 Distinguish the Heterogeneity of Antibody Interactions with Plasmodium falciparum 633 Circumsporozoite Protein Antigens. Journal of immunology 201, 1315-1326 (2018). 634

19. Bonsignori, M., et al. Maturation Pathway from Germline to Broad HIV-1 Neutralizer of a 635 CD4-Mimic Antibody. Cell 165, 449-463 (2016). 636

20. Schultheiss, C., et al. Next-Generation Sequencing of T and B Cell Receptor Repertoires 637 from COVID-19 Patients Showed Signatures Associated with Severity of Disease. 638 Immunity 53, 442-455 e444 (2020). 639

21. Yuan, M., et al. Structural basis of a shared antibody response to SARS-CoV-2. Science 640 369, 1119-1123 (2020). 641




22

22. Rogers, T.F., et al. Isolation of potent SARS-CoV-2 neutralizing antibodies and 642 protection from disease in a small animal model. Science 369, 956-963 (2020). 643

23. Robbiani, D.F., et al. Convergent antibody responses to SARS-CoV-2 in convalescent 644 individuals. Nature 584, 437-442 (2020). 645

24. Zost, S.J., et al. Rapid isolation and profiling of a diverse panel of human monoclonal 646 antibodies targeting the SARS-CoV-2 spike protein. Nature medicine 26, 1422-1427 647 (2020). 648

25. Brouwer, P.J.M., et al. Potent neutralizing antibodies from COVID-19 patients define 649 multiple targets of vulnerability. Science 369, 643-650 (2020). 650

26. Ju, B., et al. Human neutralizing antibodies elicited by SARS-CoV-2 infection. Nature 651 584, 115-119 (2020). 652

27. Cohen, S.A., Kellogg, C. & Equils, O. Neutralizing and cross-reacting antibodies: 653 implications for immunotherapy and SARS-CoV-2 vaccine development. Hum Vaccin 654 Immunother, 1-4 (2020). 655

28. Elshabrawy, H.A., Coughlin, M.M., Baker, S.C. & Prabhakar, B.S. Human monoclonal 656 antibodies against highly conserved HR1 and HR2 domains of the SARS-CoV spike 657 protein are more broadly neutralizing. PloS one 7, e50366 (2012). 658

29. Poh, C.M., et al. Two linear epitopes on the SARS-CoV-2 spike protein that elicit 659 neutralising antibodies in COVID-19 patients. Nature communications 11, 2806 (2020). 660

30. Wec, A.Z., et al. Broad neutralization of SARS-related viruses by human monoclonal 661 antibodies. Science 369, 731-736 (2020). 662

31. Walker, L.M., et al. Broad and potent neutralizing antibodies from an African donor 663 reveal a new HIV-1 vaccine target. Science 326, 285-289 (2009). 664

32. Wu, X., et al. Rational design of envelope identifies broadly neutralizing human 665 monoclonal antibodies to HIV-1. Science 329, 856-861 (2010). 666

33. Sok, D., et al. Recombinant HIV envelope trimer selects for quaternary-dependent 667 antibodies targeting the trimer apex. Proceedings of the National Academy of Sciences 668 of the United States of America 111, 17624-17629 (2014). 669

34. Tiller, T., et al. Efficient generation of monoclonal antibodies from single human B cells 670 by single cell RT-PCR and expression vector cloning. Journal of immunological methods 671 329, 112-124 (2008). 672

35. Doria-Rose, N.A., et al. New Member of the V1V2-Directed CAP256-VRC26 Lineage 673 That Shows Increased Breadth and Exceptional Potency. Journal of virology 90, 76-91 674 (2016). 675

36. Sarzotti-Kelsoe, M., et al. Optimization and validation of the TZM-bl assay for 676 standardized assessments of neutralizing antibodies against HIV-1. Journal of 677 immunological methods 409, 131-146 (2014). 678

37. Suloway, C., et al. Automated molecular microscopy: the new Leginon system. J Struct 679 Biol 151, 41-60 (2005). 680

38. Lander, G.C., et al. Appion: an integrated, database-driven pipeline to facilitate EM 681 image processing. J Struct Biol 166, 95-102 (2009). 682

39. Voss, N.R., Yoshioka, C.K., Radermacher, M., Potter, C.S. & Carragher, B. DoG Picker 683 and TiltPicker: software tools to facilitate particle selection in single particle electron 684 microscopy. J Struct Biol 166, 205-213 (2009). 685

40. Scheres, S.H. RELION: implementation of a Bayesian approach to cryo-EM structure 686 determination. J Struct Biol 180, 519-530 (2012). 687

688



Date post:	04-Feb-2021
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Cross-reactive serum and memory B cell responses to spike ......2020/09/22 · 103 suggests that...

Documents